ELECTIONS, GOVERNANCE, AND ACCOUNTABILITY IN · PDF filethe Chinese are culturally unfit for democracy. ... Governance, and Accountability in Rural China 127 ... Elections, Governance,

ELECTIONS, GOVERNANCE, ANDACCOUNTABILITY IN RURAL CHINA

Fubing Su and Dali Yang

While competitive elections are a crucial aspect of villagedemocracy, they are only one means to achieving good gover-nance. In this article, we consider village elections in thebroader context of improving governance in China. We arguethat the promotion of village democracy should be comple-mented by simultaneous improvements in a number of othernon-electoral areas that enhance the accountability of publicofficials. The Chinese central government has paid greaterattention to these institutional aspects in recent years. Theleadership is willing to encourage local experiments and toler-ate some unorthodox practices. This interaction between localinitiatives and central interventions will shape the future ofvillage democracy in rural China. We analyze several of themin particular: tax-for-fee reform, new accounting practices, vil-lage affairs disclosure, and institutionalized participation indemocratic decision making. This article reveals great diversi-ty and discusses the politics of these institutional changes.

Key words: democracy in East Asia, China

Introduction

Village democracy has attracted considerable attention inrecent years. Nowadays, millions of villagers head to voting sta-

ASIAN PERSPECTIVE, Vol. 29, No. 4, 2005, pp. 125-157.

_______. Organizing through Division and Exclusion: China’s HukouSystem. Stanford, Calif.: Stanford University Press, 2005.

_______. “Reformed Migration Control and New List of the Tar-geted People: China’s Hukou System in the 2000s,” The ChinaQuarterly, No. 181 (March, 2004), pp. 115-32.

Wang Taiyuan. Huzheng yu renkou guanli lilun yanjiu zhongshu(Comprehensive Summary on the Theoretical Study of theHukou System and Population Management). Beijing: Qun-zhong Press, 1997.

Yang, Dali. Beyond Beijing: Liberalization and the Regions in China.London: Routledge, 1999.

Zhong Yicai. “Chengxiang eyuan shehui de yonghe yu yingnongjingcheng” (The Merging of the Dual Urban-Rural Societiesand the Pulling of the Peasants into the Cities), Shehui kexue(Social Sciences), No. 1 (1995), pp. 55-58.

124 Fei-Ling Wang

hyekyung

_______. Organizing through Division and Exclusion: China’s Hukou System. Stanford, Calif.: Stanford University Press, 2005. _______. “Reformed Migration Control and New List of the Targeted People: China’s Hukou System in the 2000s,” The China Quarterly, No. 181 (March, 2004), pp. 115-32. Wang Taiyuan. Huzheng yu renkou guanli lilun yanjiu zhongshu (Comprehensive Summary on the Theoretical Study of the Hukou System and Population Management). Beijing: Qunzhong Press, 1997. Yang, Dali. Beyond Beijing: Liberalization and the Regions in China. London: Routledge, 1999. Zhong Yicai. “Chengxiang eyuan shehui de yonghe yu yingnong jingcheng” (The Merging of the Dual Urban-Rural Societies and the Pulling of the Peasants into the Cities), Shehui kexue (Social Sciences), No. 1 (1995), pp. 55-58. 124 Fei-Ling Wang

Village Self-governance and Elections in China

For at least the past century, there has been a long-runningdebate about whether the trappings of democracy, or competitionin the political marketplace, can sink roots in China. The travailsof the past century have in turn added to the pessimism of manyChina observers. As Steven Mosher, one of the most persistentcritics of China, recently wrote:

There is no Chinese tradition of respect for human rights, indeed,no notion of inalienable rights at all. There is no ghost of a sugges-tion that government in any way derives its just powers from theconsent of the governed. From the beginning of Chinese recordedhistory, the emperor has been an absolute despot. In this regard,there is little to distinguish Qin Shihuang, who ruled from 221-206B.C., and Mao Zedong, who ruled from A.D. 1949-1975, China’sautocratic traditions provide no roots, and precious little foothold,for the foreign shoots of democracy.2

For writers such as Mosher, the apparent willingness of theChinese people to tolerate the current birth planning policyreflects a deep passivity in the national character and suggests thatthe Chinese are culturally unfit for democracy. Such conclusionshave led Mosher and others to argue for treating China as anenemy.3

Obviously, Mosher is extreme in interpreting China’s history.Other distinguished scholars of Chinese history and thought,such as William Theodore De Bary of Columbia, have producedstudies that suggest China’s historical tradition was not one of amonolithic autocracy.4 For contemporary China, leaving the vil-lage elections aside, survey research by Tianjian Shi, Zhong Yang,and others have revealed much popular political participation inurban areas. It is within this context, and in a decade whendemocratization has captivated several social science disciples,that the village elections have become a major topic in studies of

Elections, Governance, and Accountability in Rural China 127

2. Steven Mosher, “Are the Chinese Ready for Liberty & Self-government?”The American Enterprise, vol. 9, No. 4 (July-August, 1998), pp. 50-53.

3. Steven Mosher, Hegemon: China’s Plan to Dominate Asia and the World(San Francisco: Encounter Books, 2000).

4. William Theodore De Bary, The Liberal Tradition in China (New York:Columbia University Press, 1983).

tions in the countryside and elect their leaders according to theOrganic Law of Villagers’ Committee. By the end of June 2005,nine provinces had already completed their elections. Sevenprovinces were in the middle of this process and five more wouldlaunch their own in the second half of the year. By the end of 2005,all villages will have completed two rounds of elections since thelaw was formally enacted in 1998.1

While competitive elections are a crucial aspect of villagedemocracy, they are only one means to the end of good gover-nance. In this article we consider village elections in the broadercontext of improving governance in China. We argue that the pro-motion of village democracy should be complemented by simul-taneous institutional improvements in a number of other non-electoral areas that enhance the accountability of public officials.These other areas clearly were on the minds of the drafters of theOrganic Law. According to the document, the ultimate goal of vil-lage democracy is to achieve self-governance in the Chinese coun-tryside. The law states that there are four components of villagedemocracy: democratic elections, democratic decision making,democratic management, and democratic supervision. To holdvillage leaders accountable, all these institutions need to be inplace. In fact, improvement in the three other areas tends to helpimprove the chances for sustaining sound electoral practices. Incontrast, the introduction of elections in an inhospitable environ-ment may ruin the reputation of elections, give democracy a badname, and thus harm the consolidation and further expansion ofdemocratic governance.

The article is organized as follows. We first review Chinesevillage self-governance in the past two decades. The next sectionprovides some theoretical arguments as to how elections may ormay not improve governance. Drawing on insights from the lit-erature on institutional design, we discuss some basic principlesin facilitating control and accountability. This is followed by ananalysis of a number of non-electoral institutions in recent years.The article concludes with some remaining issues.

126 Fubing Su and Dali Yang

1. “Chinese Villagers’ Committees elections Have Three Clear Featuresand Many Institutional Innovations.” See online at http://news.tom.com/1988/20050712-2297178.html.

travel limitations, the MCA officials first worked with provincescloser to Beijing and gradually fanned out spatially. As a result,Yunnan (an ethically diverse but less developed province in theSouthwest) and Guangdong (Canton, the economic powerhouseadjacent to Hong Kong) were two of the last provinces to jumpon the bandwagon of popular village elections.8 By the late 1990s,the village elections had become accepted practice nationwide;they were seen as a major accomplishment of China’s politicalreforms and socialist democracy. Building on this record, thePRC Organic Law of Villagers’ Committees was formally enactedby the NPC in November 1998.

Given the relatively small size and homogeneity of villagesand the close proximity between village residents and the candi-dates, elections (in 1999) for the 801,000 village committees inChina provide an excellent setting for basic-level political partic-ipation, if not the direct democracy that theorists such as Rousseauwould desire. Although the villages are not formally a part ofthe state apparatus, the very fact that meaningful elections aretaking place in such a multitude of villages (and, in the future,in urban neighborhoods) is a signal of political achievement.Moreover, over time the conduct of village elections has improved.The 1998 Organic Law of Villagers’ Committees specifically stip-ulated that the elections be conducted using secret ballots, directnominations, and more candidates than the number of positionsavailable. Scholarly studies have demonstrated that the villageelections have become a meaningful mechanism of accountabilityfor village governance.9 Some analysts have suggested that thevillage elections are becoming the training ground for furtherdemocratization.10


(2000), pp. 490-512; Tyrene White, “Village Elections: Democracy fromthe Bottom Up,” Current History, vol. 97, No. 620 (September, 1998), pp.263-67; and Jean Oi and Scott Rozelle, “Elections and Power: the Locusof Decision-Making in Chinese Villages,” China Quarterly, vol. 162(June, 2000), pp. 513-39.

8. There is a strong statistical correlation between a province’s distancefrom Beijing and the time taken by the province to approve a regulationfor village elections.

9. See especially Manion, “The Electoral Connection in the Chinese Coun-tryside.”

10. Jamie Horsely, “Village Elections: Training Ground for Democratization,”

China. For, until recently, various members of the elite, both with-in the ruling circles and even among anti-regime dissidents, haveargued that China is too backward and simply has too many“farmers” for democracy.5

Article 111 of the 1982 Chinese Constitution provided for theestablishment of villagers’ committees (VCs) and specified that theresidents elect the chairman, vice chairmen, and members of eachresidents’ or villagers’ committee. This provision became legallyexecutable with the passage of the Organic Law of Villagers’ Com-mittees (for trial implementation) by the National People’s Con-gress (NPC) in 1987. At the time, Peng Zhen, chairman of the NPCand one of the more orthodox party politicians, strongly support-ed the enactment of the Organic Law whereas Zhao Ziyang, theparty general secretary who would be ousted in 1989, was lessenamored of it.6 In a reflection of the somewhat uncertain politicalstatus of the village elections, the Organic Law was promulgatedfor trial implementation only—a strategy Chinese leaders hadskillfully used in the rural reforms of the late 1970s and early1980s.

Formal implementation of the Organic Law of Villagers’Committees began in 1989-1990—not the most auspicious circum-stances. Yet, armed with the imprimatur of the Constitution andthe national law, a small number of bureaucrats in one depart-ment of the Ministry of Civil Affairs (MCA) took their missionseriously and steadily promoted the village elections province byprovince.7 In fact, under the constraints of limited personnel and


5. Daniel Kelliher, “Keeping Democracy Safe from the Masses: Intellectualsand Elitism in the Chinese Protest Movement,” Comparative Politics, vol.25, No. 4 (July, 1993), pp. 379-96.

6. Tianjian Shi, “Village Committee Elections in China: Institutionalist Tac-tics for Democracy,” World Politics, vol. 51, No. 3 (1999), pp. 385-412.

7. There is a growing number of studies on the village elections, notablyKevin J. O’Brien, “Implementing Political Reform in China’s Villages,”Australian Journal of Chinese Affairs, vol. 32 (July, 1994), pp. 33-67; KevinO’Brien and Lianjiang Li, “The Struggle over Village Elections,” inRoderick MacFarquhar and Merle Goldman, eds., The Paradox of China’sReform (Cambridge: Harvard University Press, 1999), pp. 129-44; andMelanie Manion, “The Electoral Connection in the Chinese Countryside,”American Political Science Review, vol. 90, No. 4 (1996), pp. 736-48. Forrecent overviews of the issues, see Robert Pastor and Qingshan Tan,“The Meaning of China’s Village Elections,” China Quarterly, No. 162

there are two mechanisms for elections to improve the responsive-ness and control of the public office holders.13 The forward-look-ing one focuses on voters’ ability to select good politicians withsimilar policy orientations. Since every individual may have piecesof information about the candidates (especially in the villagecommunity with typically one thousand or so people), popularelections are the best mechanism to make use of that decentralizedinformation and pick the most capable or ethical leaders. In orderto get elected, candidates need to listen to the public and try topropose policies that best serve the majority of the community.This is a mechanism for aggregating scattered information andproducing sound public policies. Generally speaking, capablepeople with sound policy platforms will win the competition. Thebackward-looking mechanism emphasizes rewarding or punish-ing politicians depending on their behavior in office. If politiciansrenege on their campaign promises after taking office, they will bethrown out in the next election. Therefore, politicians have incen-tives to stick to their campaign promises.

But for several reasons, the electoral mechanism alone maynot be sufficient to keep politicians in check. First, informationasymmetry between candidates and voters may seriously under-mine the reputation mechanism. One crucial assumption of theminimalist argument is that information about the candidatesand their track records are readily available in the society, sovoters can reward or punish the candidates accordingly. For citi-zens with only bounded rationality, however, the above assump-tion requires too much cognitive capability from the voters. In


1942). See also discussions in Fareed Zakaria, “Illiberal Democracy,”Foreign Affairs, vol. 76, No. 6 (November-December, 1997), pp. 22-43, andPhilipper Schimitter and Terry Lynn Karl, “What Democracy Is . . . andIs Not,” in Larry Diamond and Marc Plattner, eds., The Global Resurgenceof Democracy (Baltimore, Md.: Johns Hopkins University Press, 1996),pp. 49-62.

13. James Fearon has a more formal treatment of this subject. See his “ElectoralAccountability and the Control of Politicians: Selecting Good Typesversus Sanctioning Poor Performance,” in Adam Przeworski, SusanStokes, and Bernard Manin, eds., Democracy, Accountability, and Repre-sentation (Cambridge: Cambridge University Press, 1999), pp. 55-97.Also important is John Ferejohn, “Incumbent Performance and Elec-toral Control,” Public Choice, vol. 50, Nos. 1-3 (January, 1986), pp. 5-25.

By building democracy from the bottom up and amongChina’s less educated, the successful popularization of villageelections offers a powerful counter-argument against those whosee agrarian China as unsuited for democracy. Yet, while the pro-cedures for village elections have seen a steady improvementover time, there is also no denying that elections per se are nopanacea for rural China’s difficulties, including lagging incomesand corruption. Moreover, some of the village elections have beenplagued by problems that are found in elections in other coun-tries, including vote buying and various forms of manipulation.

Democracy, Election, and Good Governance

From Election to Governance

The idea of democracy is to minimize the arbitrary exerciseof domination and ameliorate asymmetries of power. Publicoffice-holders should be responsive and accountable, while therights and interests of ordinary citizens are protected. In Westerndemocratic theory, there has been a long-standing debate on themeaning of democracy. Formal (or minimalist) representativedemocracy and civic-participatory substantive democracy aretwo major strands of thinking on this topic.11 They offer differentanswers to the following question: Is reelection sufficient forholding public officials responsive and accountable? While theminimalist tradition answers in the positive, the civic traditionargues that more direct participation by citizens and constantvigilance are needed to protect citizens’ rights.

Most scholars of democracy today work with the minimalistdefinition advanced by Joseph Schumpeter that democracy entailsperiodic competitive elections.12 From a theoretical point of view,


China Business Review, vol. 28, No. 2 (March-April, 2001), pp. 44-52. Foran example of township level elections, see Melanie Manion, “ChineseDemocratization in Perspective: Electorates and Selectorates at the Town-ship Level,” China Quarterly, No. 163 (2000), pp. 133-51.

11. John McCormick, “Machiavellian Democracy: Controlling Elites withFerocious Populism,” American Political Science Review, vol. 95, No. 2(June, 2001), pp. 297-313.

12. Joseph Schumpeter, Capitalism, Socialism, Democracy (New York: Harper,

public funds are squandered, and human lives may be lost. More-over, it is costly to overcome the collective action problem and ini-tiate these processes. One way of avoiding these problems is tobuild checks and balances into the decision-making process, andmake sure that no one, whether appointed or democratically elect-ed, can determine the fate of public policies without going throughproper scrutiny. This is based on a skeptical view of human naturebut it is precisely the nature of democracy.

Therefore, good governance needs not only periodic compet-itive elections but also supporting institutions that provide over-sight and checks on the public office-holders. A parallel can bedrawn from the marketization of the former socialist economies.Before the reform, it was generally believed that privatizationand competition would automatically lead to a free-market econ-omy. The events afterward, especially in Russia, have remindedscholars of the importance of other institutions for a competitivemarket system. In the case of democracy, theorists recognize thatseeking reelection alone may not be sufficient to make politiciansresponsive and accountable.16 They realize that “accountabilityagencies” may be needed to supplement the electoral mechanismand control politicians. These agencies include independent elec-toral commissions, independent accounting offices, and indepen-dent statistical agencies. These institutions, in combination withelectoral processes, make up the tapestry of modern democraticgovernance.

In the real world, plenty of examples support this claim. TheUnited States, despite a century of democratic elections, founditself in deep crisis by the latter half of the nineteenth century.Governance was in shambles, and rampant corruption, gangviolence, irregular markets and major accidents plagued society.This provoked a series of institutional innovations and started thestate-building processes that are now known as the ProgressiveEra. As a result of these efforts, a regulatory state emerged.17

These developments saved its democracy from collapsing.Likewise, many new democracies have found themselves


16. Ibid.17. Steven Skowronek, Building a New American State: the Expansion of National

Administrative Capacity, 1877-1920 (Cambridge: Cambridge UniversityPress, 1982).

addition, public office holders have an advantage in this reputa-tion game. Without proper checks, they can create more noise inthe information dissemination process or even mislead the votersby bending information in their favor. Some supporting institu-tions need to facilitate the flow of information and ensure itsauthenticity; otherwise deviations are not likely to be promptlydetected and punished.

Second, democratic elections are inherently uncertain.14 Eventhough in the long term and on average, good and capable peoplewill prevail through competition in the political marketplace,voters may make mistakes or evil politicians could successfullymimic good persons. This is especially problematic for newlydemocratized countries. It usually takes time for voters to learnthe rules of the game and vote responsibly. In addition, the timehorizon of politicians varies greatly. While future reelection maybe enticing enough to keep some politicians honest and respon-sive, others may have a shorter time horizon, due to an idiosyn-cratic personality or a dim prospect of reelection. The reelectionmechanism will not deter these latter kinds of officials, in whichcase it is dangerous to entrust much power to them.

Third, and closely related to the above point, there may beinstitutions to correct these problems, but post-facto correction canbe too costly in certain instances. Stokes has demonstrated howsome Latin American leaders adopted policies that went againsttheir mandates after being democratically elected. Of course, inher cases, the politicians were believed to be fighting for the long-term economic welfare of the country.15 But it is conceivable andcommonplace to see democratically elected public officials dosomething contrary to the majority interest. There are mechanismssuch as impeachment and votes of no-confidence to throw themout of office. But resorting to such mechanisms requires much timeand cost. By the time the process is initiated, the harm is alreadydone. Sometimes, the damage is irreversible: Investments go sour,


14. Przeworski even defines genuine democratic election with uncertainty.See Adam Przeworski, Democracy and the Market: Political and EconomicReforms in Eastern Europe and Latin America (Cambridge: CambridgeUniversity Press, 1991).

15. Susan Stokes, “What Do Policy Switches Tell Us about Democracy,” inPrzeworski et al., eds., Democracy, Accountability, and Representation, pp.98-130.

and control problems in a hierarchical setting. The most standardsetup is a principal-agent model, which is complicated by controlproblems arising from information asymmetry and interest diver-gence. Scholarly analyses have sought to identify institutionalmechanisms that will not only keep agents in compliance but alsoimprove overall efficiency. Therefore this literature seems to beparticularly relevant to governance.

Some basic principles that emerge from the institutionaldesign literature include ex ante control, ex post control, cost-efficiency, transparency, professionalism, and checks and bal-ances. As we will now discuss, these concepts and principles areuseful for understanding the rationale as well as the weaknessesof some institutional innovations in the Chinese countryside.

While monitoring and punishment can deter agents fromshirking, it is also very costly for the principal since the harm isalready done when punishment is meted out. A more effectivecontrol is to stamp out possible deviations ex ante. Specifically,scholars argue that administrative procedures, such as appoint-ment, budgets, and operating rules, achieve precisely this goal.21

Principals can control agents because these procedures haveeither screened out agents with different ideological preferencesor narrowed the discretion exercised by the agents. Operatingrules, in particular, stipulate as clearly and with as much detailas possible what each agent should do and should not do underdifferent situations. The U.S. government owns and manageshundreds of acres of forests employing thousands of forestrangers. As part of the control method, forest rangers are requiredto follow the manual (two volumes containing thousands ofpages) strictly, and to keep a diary of their daily activities. As aresult, despite the tremendous centrifugal forces in the bureau-cracy, the level of compliance is remarkable.22 These administra-tive procedures have in essence reduced the agents’ discretion,


Levi, Kenneth Shepsle, Barry Weingast, and Matthew McCubbins.21. The most cited piece is Matthew McCubbins, Roger Noll, and Barry

Weingast, “Administrative Procedures as Instruments of Political Con-trol,” Journal of Law, Economics, and Organization, vol. 3 (1987), pp. 243-77. In the American context, scholars are divided as to whether thepresident or the Congress really controls the bureaucracy.

22. Herbert Kaufman, The Forest Rangers: A Study in Administrative Behavior(Baltimore, Md.: Johns Hopkins Press, 1960).

“haunted by old demons that they had hoped to exorcise withdemocratic rule: violations of human rights, corruption, clien-telism, patrimonialism, and the arbitrary exercise of power.”18 Aswe look around the world in the early twenty-first century, Russiahas become a basket case of corruption and bad governance. Thedemocratically elected president of the Philippines was forced toresign for corruption while the Indonesian president, who came topower democratically following the overthrow of the Suhartoregime, was charged with corruption. And it is not just newdemocracies that must confront the scourge of corruption. Indiais caught in a massive corruption scandal concerning defenseprocurements. There are major corruption probes in Germany andFrance. The list goes on and on.

With respect to village elections in China, we find that a wor-risome sign is already emerging in rural areas. In some villages,elections without proper institutional support have indeed ruinedthe reputation of democracy in some places. Even in a widelypublicized model of democratic election, Lishu county of Jilinprovince, villagers are cooling down and some have becomesuspicious of the electoral process. When asked why he had lostinterest in elections, one villager responded, “it is meaningless.Whoever gets elected will cash in their power.”19 As a result, elec-tion turnout has moderated and even declined in some areas.

From Theory to General Principles

Democratic theory provides philosophical underpinnings butfalls short in supplying concrete proposals for institution building.For this purpose, we turn to the institutional design literature.Growing out of transaction cost theories in economics, this litera-ture finds their most fruitful applications in international politicaleconomy and American politics, especially studies of the Ameri-can congressional structure and legislature-bureaucracy rela-tions.20 Scholars in this tradition place their emphasis on oversight


18. Andreas Schedler, Larry Diamond, and Marc Plattner, eds., The Self-Restraining State (Boulder, Colo.: Lynne Rienner, 1999).

19. Zhang Jing, Problems of Rural Level Governance in China (Hangzhou: Zhe-jiang People’s Press, 2000), p. 209.

20. For some standard references, see the works of Douglas North, Margaret

flows, transparency can also significantly lower the costs for themonitoring efforts by the principal or by third parties.

While the material incentive of agents can be shaped by var-ious control mechanisms, people are also susceptible to non-material controls. Lindblom argues that persuasion is one ofthree means to control humans, the other two being power andexchange.25 Recently, more material-oriented scholars haveargued for the usefulness of ideology and corporate cultures inobtaining compliance within organizations. Professionalism notonly utilizes material incentives, but also appeals to humanbeings’ psychological needs. Through training and various sym-bols, agents can be assimilated into the profession with whichthey are associated. They take pride in being a member and tryto stick to the ethics of the community. Material losses aside,non-compliant agents also develop a great deal of shame if theyare caught.

No human is perfect and no rules will operate flawlessly allthe time. Therefore, good governance should avoid placing toomuch power in any one institution and create a governmentwith human weaknesses in mind. This point was always close tothe hearts of the drafters of the U.S. Constitution, particularlyJames Madison, who was also one of the authors of the Federal-ist Papers. Those men deliberately divided the state apparatusinto three branches, endowing each with the authority to coun-terbalance the others. Such an arrangement entails significantinstitutional costs, but it is worthwhile as an attempt to guaran-tee vital public interests. In the case of governance at the villagelevel in China, the problem is further complicated by the factthat most village committees consist of only three to seven mem-bers. That means they can easily collude to infringe on the vil-lagers’ interests, especially in financial matters.

The principles discussed above will not lead to foolproofstrategies, but they offer a perspective that we can use to evalu-ate the design of various institutions. They can also shed lighton possible institutional remedies.


sentation, pp. 131-53.25. Charles Lindblom, Politics and Markets: the World’s Political-Economic

Systems (New York: Basic Books, 1977).

leaving little room for abuse.While administrative procedures can mitigate the control

problem significantly, monitoring and punishment are neverthe-less necessary. Screening cannot be perfect, and “bad” personsmay successfully fool the superiors. Moreover, the operatingrules (like sunshine provisions) make defections more visible;but some enforcement is needed to ensure adherence by theagents. These mechanisms include self-reporting by the agents,inspecting, rewarding the loyal, and punishing the deviants.Control mechanisms should be constantly in place to make themeffective.

Any control comes at a price and good institutional designshould try to minimize the costs associated with the monitoring.One clear alternative to monitoring by the principal is to shiftthe burden to a third party. Ideally, the principal will spendvery few resources on monitoring the agents, and only respondto complaints filed by an injured party. The specific examplethat receives most academic attention is the fire alarm mecha-nism.23 A truly independent media can also serve that purpose.For a principal with very limited resources, this is an attractivealternative.

The fundamental problem with effective control is informa-tion asymmetry. If the principal knew what the agents haddone, the control problem would simply disappear. Therefore,information disclosure is essential to control, and it is necessaryfor both electoral and non-electoral mechanisms to function. Asdiscussed earlier, for the electoral mechanism to work, publicofficials should develop a reputation or generate a track recordof their past activities. Many institutional rules, such as reveal-ing the voting records in the legislature, provide just such valu-able information for the voters.24 By facilitating information


23. For an application to the Chinese context, see John Brehm and EmersonNiou, “Police Patrol vs. Self-Policing: A Comparative Analysis of theControl Systems Used in the Former Soviet Union and CommunistChina,” Journal of Theoretical Politics, vol. 9, No. 1 (January, 1997), pp.107-30.

24. Based on this logic, John Ferejohn argues that legislators in the U.S. Con-gress opted for disclosing the voting records of all legislators. Ferejohn,“Accountability and Authority: Toward a Theory of Political Account-ability,” in Przeworski, et al., eds., Democracy, Accountability, and Repre-

Even though closely related, revenue collection and publicexpenditure can be treated somewhat differently in the analysis.We will mainly focus on revenue collection in this section anddiscuss some institutional solutions.

Many peasants are complaining about the widespread col-lection of illegal fees by village cadres. In fact, peasants’ burdenshave increased quite rapidly since the early 1990s. This has ledto an increased number of visits to local Letters and Visits offices(Xinfang) by disgruntled peasants and, in some cases, large-scaleviolent confrontations.29 The causes of the peasants’ burdens aremany, but the central government is partly to blame. In 1994,China overhauled its old fiscal system and adopted a separate-tax system (fen shui zhi). There were two major goals of thischange. The first was to increase the percentage of governmentrevenue in total GDP, and the second was to allocate more rev-enue to the central government. The reform was reasonably suc-cessful but it also led to a fiscal crisis at the local level.30 Facedwith a tight budget and increasing public expenditures mandatedby the central government, many local governments felt a finan-cial pinch. County and township governments in poor areaswere especially hard pressed to find new sources of income, andthat usually meant creating new categories to collect moneyfrom peasants. The central government found itself in an awk-ward position. If it wanted to get things done locally, it had totolerate this local practice. Of course, like most public officialsworldwide, local government officials, especially township andvillage cadres, abused their discretionary power. They createdmany new ways to raise revenues and wasted the money onextravagant wining and dining, entertainment, and luxury cars.Because of their assumed discretion, local cadres, regardless ofwhether they were appointed or popularly elected to their posts,all faced a strong temptation to misuse their power. The centralgovernment has been experimenting with policy measures in


Lu, “Taxation Without Representation: Peasants, the Central and theLocal States in Reform China,” China Quarterly, vol. 163 (September,2000), pp. 742-63.

29. See ibid. for some examples.30. For more discussion, see ibid. and Ran Tao and Mingxing Liu, “Regional

Competition, Fiscal Reform and Local Governance in China” (workingpaper, 2004).

Non-Electoral Institutions and Accountability in Rural China

The Chinese central government is now putting more empha-sis on promoting non-electoral institutions. The most recent pol-icy regarding village democracy (No. 17 [2004]), for example,specifically clarifies issues related to transparency and democra-tic management. It also tries to standardize rules and practicesin the country.26 As in previous policy changes in rural China,the top leadership has been fairly open-minded so far: It is will-ing to allow local governments and officials to experiment withalternative ways of rural governance. The Central Party Schooland the Central Editing and Translation Bureau (Zhongyang BianyiJu), together with Beijing University, have even set up a prize forinstitutional innovation by local governments, and some prize-winners have created new village governance models.27 Despiteits official status, this prize is valued by university professorsand therefore has a high degree of independence and credibility.The central government’s tolerance is a clear indication of itsflexibility, which will encourage further local experiments. Ofcourse, the central leadership is not entirely passive in thisprocess. It listens carefully and promotes good innovations whilevetoing others. These choices can enhance village self-gover-nance in the countryside but also reflect the party’s own politicalinterests. Undoubtedly, these dynamic interactions between cen-tral intervention and local innovations will determine the evolu-tionary path of rural governance.

Tax-for-Fee Reforms and the Abolishment of the Agricultural Tax

No other area is more prone to official abuse than villagefinance. And this is indeed the major source of rural instability.28


26. General Office of the CCP Central Committee and General Office of theState Council, “Opinions on Promoting and Perfecting Village AffairsPublicity and Democratic Management” (June 22, 2004), online athttp://news.xinhuanet.com/zhengfu/2004-07/12/content_1591421.htm.

27. A full list of prize-winning institutional innovations is available atwww.chinainnovations.org. The third round of competition is currentlyunder way.

28. For more analysis of this phenomenon, see Thomas Bernstein and Xiaobo

tions to answer queries from peasants and thus ensure thatfarmers paid taxes with a full understanding of what they werepaying for, thereby reducing the chances of local officials collect-ing unauthorized levies.34 In 2001, Jiangsu and Zhejiang joinedthe experiment, and the program further expanded to twentyother provinces in 2002.

In 2003, the central government decided to promote the tax-for-fee reform nationwide.35 Under this new system, local govern-ments were allowed to collect only one agricultural tax from peas-ants and no fees were allowed to be charged. Legal charges fortownship governments, such as the so-called wu tong (for ruraleducation, family planning, militia, support for veterans, androad construction), were incorporated into the agricultural tax.The san ti in villages (for collective investments, welfare, andcadre compensations) was added as a surcharge. As a result, theagricultural tax rate was raised from 3 percent to 7 percent, and a20-percent surcharge was added for village level finance. In total,peasants would pay no more than 8.4 percent of their income.Additionally, tax collectors from local taxation agencies would bein charge of the collection, and township and village cadres couldhelp but were not in direct control.

In early 2004, Premier Wen Jiabao announced at the NPC’sannual meeting that the state was planning to phase out theagricultural tax in three years.36 This may be based purely on acost-benefit analysis of tax collection. Many officials acknowl-edge the high cost of collecting taxes in vast and remote ruralareas. By February 2005, twenty-six provinces had already abol-ished the agricultural tax and by 2006 this tax category will


34. CCTV, January 24, 2001.35. “State Council’s Opinion About Promoting Rural Tax-for-Fee Reform

Trial Work in the Country,” March 27, 2003, online at www.China.org.cn/Chinese/Pl-c/331156.htm.

36. In addition to alleviating peasants’ burdens, there is also a simple reasonfor abolishing agricultural tax: It makes economic sense. Because ofChina’s highly scattered population, it is very costly to collect taxesfrom many individual households. Local governments often employextra manpower to collect taxes, but the revenues can barely cover thecosts. For some concrete examples, see Yang Weijun, “A Study of theImpact of Abolishing Agricultural Tax on Grassroots Governments,”online at http://guancha.gmw.cn/showaspx?id=3066.

the past few years, one thrust of which is to curb local cadres’discretion.

One such major change is the tax-for-fee reform. In the early1990s, the Ministry of Agriculture, in response to the problem ofpeasants’ burdens, proposed a unitary tax in rural areas. Peas-ants who were affected expressed strong support for the idea.31

According to this plan, local cadres would not have the discre-tion to levy fees on peasants, but could only collect one nation-wide agricultural tax from peasants. But the proposal was puton hold because the country was caught in the growth euphoriathat followed on the heels of Deng Xiaoping’s celebrated south-ern tour in 1992. As growth began to slow down, however, somerural localities began to experiment with tax reforms. Beginningin 1995, Wugang (Hubei province) was authorized to experimentwith tax-for-fee reforms in rural areas. By standardizing fees intotaxes, Wugang’s experimental townships were able to reduceburdens on farmers as well as reform public finance.32 In Anhuiprovince, one of the pioneers of rural reforms in the post-Maoera, officials in Taihe and Woyang counties initiated a similarexperiment in 1994 with the approval of provincial authorities.33

Taihe officials sought to rein in arbitrary levies on peasants whowere caught between stagnant prices for farm products and ris-ing costs. By 1999, fifteen counties (cities) had adopted rural taxreforms.

Based on local trials, the central government chose Anhui toexperiment with the tax-for-fee reform province-wide. For thefirst year of the reforms, the financial burdens on peasants inAnhui were reduced by 1.7 billion yuan, or 31 percent. With theimplementation of the reforms, per capita tax and levies decreasedto 75.5 yuan. To boost the transparency of the reforms, the coun-ties also set up supervision stations staffed by members of thecounty people’s congresses. Tax administrators also set up sta-


31. Dali Yang, Calamity and Reform in China (Stanford, Calif.: Stanford Univer-sity Press, 1996), p. 210.

32. China Economic Times, June 24, 1999.33. The practice was started in 1993 in Xinxing Township of Anhui’s Woyang

County, against the express prohibition of the county people’s congress.The Xinxing leadership fixed the amount of farmer taxes and fees at 30yuan per mu. For details of this case, see China Economic Times, February24, 2001.

On the positive side, the tax-for-fee reform has tied local cadres’hands and forced them to downsize bloated local governments.To raise funds, village leaders in particular must seek consentfrom villagers and go through villagers’ assemblies (cunmin dahui)or villagers’ representative assemblies (cunmin daibiao dahui).

Accounting and Auditing Reforms

It is clear from various reports on peasants’ petitions (shang-fang) and other media investigations that the most contentiousissue in peasant-cadre conflicts in the villages is financial mis-management and corruption. Peasants have accused villagecadres of corruption, embezzlement, and misuse of public fundsto support lavish lifestyles. Some surveys further disclose thatmost villages do not follow basic accounting practices. Accord-ing to one survey, 90 percent of the villages in the sample do nothave formal accounting rules and regulations and have noaccounts for collective assets. Seventy percent of village accoun-tants are unclear about their responsibilities and the limits oftheir discretion, and half of the accountants also serve as cashiers.Indeed, 20 percent of villages do not even have accountingbooks and 30 percent of accountants refuse to pass their bookson to their successors.43 In some cases, the village leader (eitherthe party secretary or villagers’ committee director) also servesas the accountant and the cashier. More often the accountant iseither a relative or close ally of the village leader.

The backwardness of village financial management hasmade village-level finances a hotbed of corruption and misman-agement, but it also means that certain basic reforms, such as thepopularization of sound accounting practices, could generatelarge payoffs. One institutional development entrusts the man-agement of village accounts to township governments (cunzhangxiangguan or cuncai xiangjian). Under this arrangement, thetownship government establishes an accounting service officeand is responsible for hiring the accountants. All villages in thejurisdiction transfer their accounts to the accounting office andpay fees for its services. Every village still needs one cashier andis to have full control of the collective funds in its accounts. This


43. Zhongguo shenji bao (China Auditing News), November 21, 2000.

finally disappear in China.37

From the perspective of controlling local officials, the tax-for-fee reform will clearly lead to better governance. In essence, thisreform has curbed the discretionary power of township and vil-lage cadres. By centralizing revenue collection in tax agencies inhigher-level governments, the central government is signaling itsunwillingness to tolerate “creative” fee collection. While holdinglocal cadres more accountable, this policy change does imposemore financial hardships on local governments and villages and itcompromises their ability to provide public services. Abolishingthe agriculture tax, in particular, may cause already heavilyindebted local governments that were previously financed largelyby agricultural surpluses to crash. For example, with the intro-duction of the tax-for-fee experiment in Anhui, township govern-ments in Anhui province had a budget of 4.6 billion yuan in 2000,but the payroll alone amounted to 4.95 billion yuan. An additional2.28 billion yuan was needed to fund rural education.38 To partlyaddress this shortfall, the central government has pledged to sub-sidize about 10 billion yuan annually to agricultural counties andtownships.39

Some provinces also began holding provincial and municipalgovernments responsible for financing rural education.40 Despitethese efforts, it is still possible that cash-strapped township orvillage cadres may fall back on fee collections again.41 The StateDevelopment and Reform Commission completed a sixth roundof inspection in mid-2005 and disclosed 13,000 cases of violationsin the rural areas, most of which were illegal fees for education.42


37. “Abolishing Agricultural Tax Nationwide and Lifting the Curtain forComprehensive Reform in the Countryside,” March 11, 2005, online atwww.lianghui.org.cn/chinese/zhuanti/2005lh/808939.htm.

38. China Economic Times, March 6, 2001.39. “Abolishing agricultural tax nationwide and lifting the curtain for com-

prehensive reform in the countryside,” March 11, 2005, online at www.lianghui.org.cn/chinese/zhuanti/2005lh/808939.htm.

40. An interview with Wang Chaowen, deputy director of the FinanceResearch Institute, Ministry of Finance, online at www.cncitizen.org/article.php?articleid=1416.

41. Ray Yep, “Can ‘Tax-for-Fee’ Reform Reduce Rural Tension in China?The Process, Progress, and Limitations,” China Quarterly, vol. 177(March, 2004), pp. 42-70.

42. Zhongguo qingnian bao (China Youth News), August 5, 2005.

administrative expenditures.47 In the past, village party secre-taries monopolized the power of village spending. With theintroduction of these financial management groups, spending inthe village will be reviewed by the small groups.

Interestingly, this institutional design was not in the OrganicLaw in 1998. After local experiments, the central leadership decid-ed to promote this institution in 2004. According to the “Opin-ions on Promoting and Perfecting Village Affairs Publicity andDemocratic Management” (No. 17 [2004]), villagers’ financial man-agement small group members are to be elected in the villagers’assemblies or villagers’ representative assemblies. The smallgroups are directly responsible to the assembly, not the villagers’committee or the party committee. The intention of creatingchecks and balances is very clear.48

Rules quickly become meaningless if violations go unpun-ished. Frequent monitoring is an integral part of any properlyfunctioning institutional design. In the financial realm, auditsplay this important role. Ideally, an independent third partyshould be invited to audit the village accounts regularly. It wasreported that Liji villagers in Weishan County, Shandong province,hired an accounting company, Weishan Xingheng Ltd., to auditthe accounts of their village leaders in 2000.49 However, thismay not be feasible for most villages. While some accountingand auditing services are licensed to practice in the metropolitanareas, their reach to the rural world remains limited. In their stead,township governments have filled the void and played an activerole. Baodi County of Tianjin has ordered township governmentsto audit village accounts on a regular basis, making the first weekof every January and July the “auditing week.” The county partydisciplinary organization and auditing bureau were authorizedto issue “auditing notice.”50 By the end of 2000, 79.3 percent of


47. Zhongguo gaige bao, November 11, 2000.48. “Opinions,” June 22, 2004, online at http://news.xinhuanet.com/zhengfu/

2004-07/12/content_1591421.htm.49. Nanfang zhoumo (Southern Weekend), Internet version, February 22,

2001. Of course, the interesting twist of this story is that the townshipgovernment interfered in the disclosure of these auditing results, andthe villagers finally had to take the problem to court.

50. Renmin ribao (People’s Daily, Beijing), August 15, 2000.

approach has been widely adopted in Hanzhuang Township ofPinggu County in Beijing and several counties in Zhejiangprovince.44 One variation of this mechanism has been practicedin localities such as Jiuguan township of Nintai city, Shandongprovince.45 It is called the accountant secondment system (kuaijiweipai zhidu), which means the township government is respon-sible for recruiting accountants through competitive exams. Theaccountants are then dispatched to the villages but superviseddirectly by the township operations station (jingying zhan) ratherthan the village leaders.

These arrangements have different names in other places,but they all share two features. First, by aligning the interest ofthe accountants more closely with township governments, theymake it more difficult for village leaders and the accountants tocollude. Second, the township intervention may help isolateaccounting from village politics, and may breed a sense of ser-vice and professionalism through more training and better pay.Both are helpful in checking the corruption and misuse of thepublic funds by village leaders.

Besides centralization and professionalization of villageaccounting, some localities have begun to allow more publicinvolvement in matters of financial management by forming vil-lagers’ financial management small groups (cunming licai xiaozu).The small groups consist of representatives elected through thevillagers’ representative assembly and are empowered to super-vise the details of village finances. They examine originalreceipts submitted for reimbursement, verify changes in collec-tive assets, determine the salaries of the village leaders, andmonitor contract enforcement and cash flows.46 For example,every village in the previously mentioned Hanzhuang Town-ship established a village financial management small group in2000. Members range from six for big villages to three for smallones. They verified all the expenses village leaders submitted forreimbursement to the township accounting office. Such financialoversight helped curtail dining and entertainment expenses and


44. Zhongguo gaige bao (China Reform News), November 11, 2000; ChinaEconomic Times, July 19, 1999.

45. Fazhi ribao (Legal Daily), October 8, 1999.46. Zhang Jing, Problems of Rural Level Governance in China, pp. 157-58.

cadres still enjoy an information advantage. An alternative way isto take advantage of villagers’ information about local conditionsand invite them to monitor their own cadres. Typically, villagersknow best what public works are needed, how much materialsreally cost, and whether or not cadres live a luxurious lifestyle.In addition, there are village affairs other than financial spending,such as the implementation of family planning policy, distributionof emergency and poverty funds, allocation of housing land,and renting and selling of collectively-owned land. All theseissues are of great concern to villagers and are vulnerable to theabuse of power. Forcing village cadres to disclose all this infor-mation can empower villagers to monitor their leaders, whichimproves transparency in village governance.

The central government started to promote village affairspublicity (cunwu gongkai) nationwide in April 1998. In a key poli-cy directive, it asked all villages to disclose major issues relevantto the interests of villagers, including new projects, collectiveassets and finance, land uses, family planning, fees and levies,contracts, and cadres’ salaries. The emphasis was on full disclo-sure of village finances.55 Following a major promotion drive,about 80 percent of the villages across the country had adoptedthe village affairs disclosure system by the end of 1999.56 Someplaces have produced specific operational rules governing suchdisclosures. For example, Laiyang city in Shandong provincedirected every village to publicize village affairs twice a year(June 25 and December 25) and financial affairs four times a year(at the end of each quarter). Every village was required to set uppublicity boards (one for villages with fewer than 200 peopleand two for those with more than 200). The information wasalso disseminated via broadcasts, village meetings, and flyers(mingbai zhi). Cadres were required to respond to questions with-in two days.57

The expectation that spending will be publicized tends tomoderate spending by village cadres and to reduce wastefulexpenditures. By the end of 1998, 93.2 percent of the counties inAnhui province had adopted village affairs transparency. Accord-


55. Economic Daily, June 11, 1998.56. Renmin ribao, November 25, 1999.57. Nanfang zhoumo, December 25, 2000.

villages in the county had been subjected to such audit.51 The 2004“Opinions” also mandates auditing of village leaders beforeelections. County and township governments are held responsiblefor this task.

From a control perspective, entrusting village accounts toaccounting service offices under township government and audit-ing villages by township governments are only second-best insti-tutional solutions. They should be replaced by more independentthird parties in the future. After all, it may be difficult to convincevillagers that township governments are truly neutral in villageaffairs. Moreover, these institutional designs, though technical,have more fundamental political implications. If villages are trulyself-governing, should township and county governments man-age village affairs? Opinions seem to be particularly divided onthe issue of “cunzhang xiangguan.” Some scholars regard this prac-tice as a violation of village self-governance and therefore shouldbe banned. Zhan Chengfu, a senior official in charge of villageself-governance in the Ministry of Civil Affairs, publicly endorsedthis view.52 But other government agencies, such as the Ministryof Finance, are more motivated by corruption control, and activelypromote county governments’ managing township accounts(xiangcai xianguan).53 Many local governments seem to be willingto sacrifice self-governance for the sake of financial control. Atleast two provinces, Fujian and Zhejiang, and some local govern-ments in Henan and Hubei provinces, have promoted cuncaixiangguan.54 This may be a worrisome sign for many people withvillage self-governance in mind, but it does reveal that improvingaccounting practices is a high priority among current leaders.

Transparency and Cunwu Gongkai

While centralizing accounts and intensifying auditing cancontrol village cadres, these are very costly practices, and village


51. Economic Daily, November 20, 2000.52. Zhan Chengfu, “Be Cautious in Promoting Cuncai Xiangguan,” Gaige

neican, No. 18 (1998).53. People’s Daily (Internet ed.), July 13, 2005.54. Xinanjiang nongwang, June 3, 2003; Renmin ribao, June 28, 2005; Zhongguo

jijian jiancha bao, June 8, 2005.

below.61

Transparency and village affairs disclosure can be a power-ful tool in facilitating information flow and control. But there arealso some serious weaknesses in the current design of this policy.One key problem remains in the shape of information asymme-try. How can villagers verify the truthfulness of the disclosedinformation? The village affairs publicity supervision smallgroup is intended to empower villagers institutionally and allowthem to access the same information as village cadres. In reality,researchers and field workers discovered that in many placesgroup members were either close associates of village cadres orsoon colluded with them.62 Second, even when group membersexercise their power, there is no institutional guarantee of satis-factory results. When village cadres refuse to disclose informa-tion, or doctor the records, group members report this to town-ship governments according to the Organic Law.

But the township party committees and governments arehardly neutral third parties in rural governance. They have incen-tives to side with village cadres, rendering this institution stillvery weak. The supervision small group should be able to issue amore credible threat to non-compliant village officials. For exam-ple, it should be given the power to call a villagers’ representativemeeting to impeach the leaders. Even this authorization mayencounter problems, since the party is still strong in village gover-nance and is not subject to villagers’ consent. This is a more fun-damental challenge and we discuss it in the next section. Becauseof the weaknesses in the institutional design, the national cam-paign has made only moderate progress. In its first national meet-ing in 2003, the coordination group for village affairs publicityadmitted that, among about 680,000 villages in China, 95 percenthad promoted publicity and only 60 percent had followed thecentral government’s instructions.63 That number does not sayanything about the quality of disclosure. In one study, only 27.7percent of villagers said they could trust the content of the public-ity boards.64


61. Renmin ribao (Huadong Xinwen), March 27, 2001.62. Fazhi ribao, January 12, 2002.63. Xinhua Net, June 13, 2003.64. It was conducted by the State Council Development and Research Center

ing to one estimate, village expenses decreased by 25.8 percentthat year.58 In Zhejiang province, where 98 percent of the villageshad achieved financial transparency, entertainment and diningexpenses by village cadres dropped by 10 to 40 percent.59 Thepromotion of transparency also has the added benefit of exonerat-ing village cadres from unfounded charges, thus helping easestrained cadre-farmer relations. One of the reasons for villagers topetition officials at upper levels is lack of transparency. Not infre-quently villagers’ suspicion and discontent feed on the secrecysurrounding village decision making and financial management.Some cadres have thus welcomed the drive toward transparencyand claimed that it cleared their names of suspicions and innuen-does of misdeeds.60

On the basis of initial successes, the central governmentformed the coordination group for village affairs publicity andlaunched nation-wide campaigns to implement this change inrecent years. More specific details were added to this policy inthe 2004 “Opinions” cited previously. Most notable is the intro-duction of one institution: the village affairs publicity supervi-sion small group (cunwu gongkai jiandu xiaozu). All members areto be elected in the villagers’ assemblies or villagers’ representa-tive assemblies. Like the villagers’ financial management smallgroup members, they are directly responsible to the villagers’assembly and supervise the villagers’ committees. In many places,the membership of these two small groups overlaps.

While the transparency measures mainly serve as an ex postcontrol mechanism, with some modifications it can also controlthe village cadres’ behavior ex ante. For example, Shandong’sJiaonan city inaugurated the “housing land approval disclosuresystem.” Villagers who want to build houses need to get approvalsfrom both village cadres and village residents. Decisions by thecadres are not final until three days after the public disclosure.Within three days, villagers may file complaints with the cityland resources bureau. Such ex ante control involves villagers inthe decision-making process, which will be further discussed


58. China Economic Times, November 23, 1998.59. Renmin ribao (Huadong Xinwen), November 8, 2000.60. Zhongguo gaige bao, November 23, 1998; Renmin ribao, June 22, 1999 and

July 19, 2000; Guizhou ribao, October 7, 2004.

assembly. This institution improved the effectiveness of villagedemocracy and was adopted by the central government in theOrganic Law in 1998.65 Now, all major issues in the village, suchas renting collective land, raising debts, and determining thesalaries of village cadres, must be discussed in the villagers’assemblies or villagers’ representative assemblies. The majoritywill prevail in the development of policies.

Public involvement can lead to better policies and morechecks on cadres’ discretion. For example, leaders at Nanlou vil-lage in Zhengding County, Hebei province, wanted to invest650,000 yuan and build a business center next to the highwaypassing by the village. When the proposal reached the villagers’representative assembly, representatives pointed out that truckswould not stop for this isolated building if no other supportingservices were around. Instead, they suggested an oil-extractingplant, which could take full advantage of the rich sideline prod-ucts in the neighboring areas. This counter-proposal was finallyaccepted and villagers even volunteered their labor for this pro-ject, which was quite unusual. The plant turned out to be prof-itable.66 Lishu county of Jilin province pioneered the open ballotelection (hai xuan) in village elections. Villager participation indecision making in this county had also been active. Villagersnot only vetoed proposals that lacked sound economic justifica-tions, but also exercised moderate checks on the fees and leviesfrom higher level governments. Village cadres cannot collectthose fees without the assembly’s authorization.67

Tax-for-fee reform and the abolishment of agricultural taxeshave created more demand for functioning assemblies. As aresult of these changes, village cadres can no longer raise fundsfor public projects. But it is also important that local communi-ties do have the financial resources to provide basic publicgoods, such as the paving of roads and the construction of nec-essary irrigation facilities and infrastructure. According to thenew rules from the central government, village leaders mustrequest permission from the villagers’ assembly or villagers’


65. Cheng Tongshun, Studies on Village Political Development in ContemporaryChina (Tianjin People’s Press, 2000), p. 186.

66. China Economic Times, July 6, 1998.67. Zhang Jing, Problems of Rural Level Governance in China, pp. 206-7.

Participatory Decision Making

Bringing villagers into decision-making processes serves twomain purposes. First, the village as a whole can take advantageof the superior local information of the residents and make effi-cient economic and social policies. Many people, including somescholars, believe that peasants are not sufficiently educated tomake good policy choices, and that the elites (party membersand village cadres) should take the lead. This is probably reason-able under certain circumstances, but when it comes to under-standing local conditions, each resident has some useful informa-tion, and the villagers as a whole have far more knowledge thana handful of elites from out of town. Secondly, villagers’ partic-ipation, adopted widely and routinely, can put some kind of exante institutional checks on village cadres. Making villagers’ par-ticipation and majority votes a necessary part of a decision-mak-ing process gives the villagers a veto right. Unlike the ex postcontrol mechanism of public disclosure, greater villager partic-ipation gives villagers a direct voice in the content of policies,thereby putting an effective control over village cadres, especial-ly in the domain of public spending. Ideally, wasteful spendingor public policies that hurt the majority interest will be vetoed atthe decision-making stage.

As in other areas, the specific institutions where villagers canexercise their decision-making power have evolved in the pasttwo decades. In the trial version of the Organic Law of Villagers’Committee, villagers should participate through villagers’assemblies (cunmin huiyi), consisting of all villagers 18 years oldor older. This is feasible for small villages of several hundred, butfor large ones with thousands of villagers, this institution is quiteclumsy. In addition, as a result of economic reform, many peas-ants migrated to cities to find jobs, so that meeting the minimumnumber of participants was a challenge for some villages. Inresponse to this practical problem, some places experimentedwith villagers’ representative assemblies (cunmin daibiao huiyi),where five to fifteen households elected one representative to the


in the late 1990s. More information can be found in Zhao Shukai, “Dilem-mas Facing Village Organizations,” online at www.chinaelections.org.The numbers reflect conditions in the late 1990s.

nized? Traditionally, the party committee was in absolute controland only in recent years have the villagers’ committees begunchallenging the party’s authority. In fact, the conflict betweenthese two committees has been one major concern for many peo-ple since the implementation of the Organic Law.

There have been some institutional innovations to solve thisconflict.72 The “two-ballots-rule” (liang piao zhi) requires that allvillagers, both party members and non-party members, vote forcandidates for village party committees, and candidates who gethalf of the votes will be formally nominated by the party for vil-lage party committees. In the second round, all party members inthe village cast a ballot to select party committee members.Another solution is called “carrying on one shoulder” (yijian tiao),where the party will nominate party committee members to runfor villagers’ committee positions. If they are elected, the samegroup of people will effectively control both committees. Thesetwo institutional innovations have the potential of solving theconflict and have been endorsed by the CCP’s Department ofOrganization. But the danger is also clear: in both cases, the partybecomes stronger and replaces the villagers’ committee. Eventhough they work through the democratic process, these changesmay harm village self-governance in the long run. Moreover, inorder to get their candidates elected, higher-level party organiza-tions have strong incentives to intervene and manipulate the elec-tion. In both cases, the villagers’ assembly or the villagers’ repre-sentative assembly is marginalized.

One recent innovation was started in Qing County, Hebeiprovince, and was widely praised by scholars and policy makers.73

Unlike the last two solutions, this “Qingxian model” centers onthe villagers’ representative assembly. The assembly is reinforcedand becomes a permanent institution. The representatives elect achairperson. At the same time, the party committee stops micro-managing village affairs and concentrates on macro issues. More


72. For interesting discussions, see Xu Fuqun, “Solutions to the ConflictBetween Two Committees,” October 7, 2004, online at www.chinaelec-tions.org; and Jing Yuejing, “Villagers’ Self-governance, Two Committees’Relations, and Chinese Political Transition,” China Reform Forum, June15, 2005, online at www.chinaelections.org.

73. Xu Fuqun, “Solutions to the Conflict between Two Committees.” Moreanalyses can be found at www.chinainnnovations.org.

representative assembly. These permissions may be granted ona case-by-case basis (yishi yiyi). In He County of Anhui province,309 villages have used this mechanism to raise money for publicworks such as irrigation stations, roads, bridges, and broadcastnetworks. The county leaders admitted that, before the tax-for-fee reform, the county had 11.22 million yuan in accumulatedpublic accumulated funds (gongjijing), or about 22 yuan perpeasant. But very little money was actually spent on peasants.With the case-by-case approval process, 7 million yuan wasraised and villagers were generally satisfied with this change.68

In June 2001, Jiaoshanpo village of Shandong province raised1,500 yuan per villager to bring the nearby river under control.The party secretary of the village noted that, if not for publicparticipation in decision making, such an undertaking wouldhave been unimaginable.69 Similar positive developments arealso reported in other villages. Villagers either vetoed over-activecadres or public consent led to speedy collection of funds.70

Of course, all these developments depend on one condition,namely, that the villagers’ assembly or the villagers’ representativeassembly can indeed function as a decision-making body. Whilepositive cases are certainly numerous, many researchers report alarge number of negative examples. Many villagers’ assembliesdid not work or villagers’ representatives were controlled orcoopted by village cadres.71 Currently, scholars and policy makersare debating the causes of this problem and looking for possiblesolutions. In a way, this touches a fundamental issue in villagegovernance, i.e., who should control village lives in rural China?Should it be the communist party, democratically elected leaders,or villagers themselves? Institutionally, how should these threeinstitutions (the party committee, the villagers’ committee, and thevillagers’ assembly or villagers’ representative assembly) be orga-


68. Xinhua Daily News, February 21, 2001.69. Banyuetan (China Comment, Internet ed.), June 24, 2001.70. Nongmin ribao, April 19, 2003; Fazhi ribao, April 7, 2004; Xiangzhen luntan,

April 2003.71. Zhongguo shehui bao, January 13, 2004; Xiangzhen luntan, November 2003;

Renmin ribao, October 22, 1997. Lily Tsai also reported that some traditionalinstitutions have played an important role in providing local publicgoods. “Cadres, Temple and Lineage Institutions, and Governance inRural China,” The China Journal, vol. 48 (July, 2002), pp. 1-27.

of village affairs. A functioning villagers’ assembly, villagers’representative assembly, active financial management smallgroup, and village affairs publicity supervision group shouldserve as a balancing force against the elected village committeemembers. They should be responsible for drafting village bud-gets, overseeing its implementation, calling for outside auditing,and initiating no-confidence processes, when appropriate. In aword, they should help solve the collective-action problem in thevillage and constantly check on the village committee and itshandful of members. This improvement can also mitigate votebuying and bribery as reported in recent years. One major con-cern regarding vote buying is that people who spend their ownmoney to get elected will abuse their power and pocket collectiveassets. Instead of electing good leaders, vote buying can sendentrepreneurs, people more interested in making money andadvancing their own interests, into village leadership positions.Institutional innovations discussed here, on the other hand, candecrease the prospect of power abuse and thus discourage thistype of candidate from seeking public office.

Given the amount of information available, we are unable toanalyze each institution in great detail. How do different regionsrespond to these opportunities? Why are some regions moreactive than others? What factors can explain the level of economicdevelopment, the degree of conflict in the villages, the local lead-ership, and the dependability of the connection to the centralbureaucracy? These are certainly important and interesting ques-tions and more empirical research is needed. Regardless of thesuccess or failure of particular institutions, however, the dynamicinteraction between local innovation and central interventionshapes the future trajectory of village self-governance. Both elec-toral and non-electoral institutions count.

Principal References

Bernstein, Thomas and Xiaobo Lu. “Taxation Without Represen-tation: Peasants, the Central and the Local States in ReformChina,” China Quarterly, vol. 163 (September, 2000), pp. 742-63.

Brehm, John and Emerson Niou. “Police Patrol vs. Self-Policing:


specifically, the party secretary must seek election for the assem-bly chair position. Secretaries who fail must resign. The villagers’committee is the operational arm of the assembly and all majorpolicies are decided in the villagers’ representative assembly.Among these new innovations, the “Qingxian model” is closestto Western-style democracy, where parties are only vehicles topower. But, from the CCP’s perspective, this may come too closeto yielding the party’s leadership position in the rural area.Therefore, despite the praise for this approach, the CCP has notendorsed this innovation. In the latest document (No. 17 [2004]),the party committee is still in a leadership position even thoughthe villagers’ assembly or the villagers’ representative assemblyhas formal power. This will certainly compromise villagers’ abili-ty to determine their own affairs. Until the CCP is willing to giveup its dominance, this kind of political struggle is likely to con-tinue in the future.

Conclusion: Improving Governance in Rural China

The popularization of village elections in China is a majorachievement of China’s political reforms and democratic devel-opment. Nevertheless, sound rural governance is far more thanthe periodic holding of elections. On the one hand, greater popularparticipation in village affairs translates into greater demand forbetter governance and is believed to have resulted in improve-ment in local administration, including greater transparency. Onthe other hand, corruption and misrule by elected village offi-cials have undermined popular trust in democratic institutionsin some localities. In this article, we have discussed some mecha-nisms for governance that have evolved in the rural areas duringthe past few years. Evidence indicates that these non-electoralinstitutions can improve the function of the democratically elect-ed village committees and facilitate a healthier development ofgood governance in rural China.

Having said that, however, is not to dismiss the value ofelections. It should be reiterated that both electoral and non-elec-toral mechanisms are important parts of good governance. Giventhe small size of the village committee, it is important to instituteregular monitoring and checks and balances in the management


162 (June, 2000), pp. 513-39.Pastor, Robert and Qingshan Tan. “The Meaning of China’s Vil-

lage Elections,” China Quarterly, vol. 162 (2000), pp. 490-512.Przeworski, Adam. Democracy and the Market: Political and Economic

Reforms in Eastern Europe and Latin America. Cambridge:Cambridge University Press, 1991.

Schmitter, Philippe and Terry Lynn Karl. “What Democracy Is . . .and Is Not,” in Larry Diamond and Marc Plattner, eds., TheGlobal Resurgence of Democracy. Baltimore, Md.: JohnsHopkins University Press, 1996, pp. 49-62.

Tsai, Lily. “Cadres, Temple and Lineage Institutions, and Gover-nance in Rural China,” The China Journal, vol. 48, (July,2002), pp. 1-27.

White, Tyrene. “Village Elections: Democracy from the BottomUp,” Current History, vol. 97, No. 620 (September, 1998),pp. 263-67.

Yang, Dali. Calamity and Reform in China. Stanford, Calif.: Stan-ford University Press, 1996.

Yep, Ray. “Can ‘Tax-for-Fee’ Reform Reduce Rural Tension inChina? The Process, Progress, and Limitations,” China Quar-terly, vol. 177 (March, 2004), pp. 42-70.

Zhang Jing. Problems of Rural Level Governance in China. Hangzhou:Zhejiang People’s Press, 2000.


A Comparative Analysis of the Control Systems Used inthe Former Soviet Union and Communist China,” Journal ofTheoretical Politics, vol. 9, No. 1 (January, 1997), pp. 107-30.

Cheng, Tongshun. Studies on Village Political Development in Con-temporary China. Tianjin: People Press, 2000.

Fearon, James. “Electoral Accountability and the Control ofPoliticians: Selecting Good Types versus Sanctioning PoorPerformance,” in Adam Przeworski, Susan Stokes, andBernard Manin, eds., Democracy, Accountability, and Repre-sentation. Cambridge: Cambridge University Press, 1999,pp. 55-97.

_______. “Accountability and Authority: Toward a Theory ofPolitical Accountability,” in Adam Przeworski, et al. eds.,Democracy, Accountability, and Representation. Cambridge:Cambridge University Press, 1999, pp. 131-53.

Horsely, Jamie. “Village Elections: Training Ground for Democ-ratization,” China Business Review, vol. 28, No. 2 (March-April, 2001), pp. 44-52.

Kelliher, Daniel. “Keeping Democracy Safe from the Masses:Intellectuals and Elitism in the Chinese Protest Movement,”Comparative Politics, vol. 25, No. 4 (July, 1993), pp. 379-96.

Manion, Melanie. “The Electoral Connection in the ChineseCountryside,” American Political Science Review, vol. 90,No. 4 (1996), pp. 736-48.

_______. “Chinese Democratization in Perspective: Electoratesand Selectorates at the Township Level,” China Quarterly,vol. 163 (2000), pp. 133-51.

McCubbins, Matthew, Roger Noll, and Barry Weingast. “Adminis-trative Procedures as Instruments of Political Control,”Journal of Law, Economics, and Organization, vol. 3 (1987), pp.243-77.

O’Brien, Kevin J. “Implementing Political Reform in China’s Vil-lages,” Australian Journal of Chinese Affairs, vol. 32 (July,1994), pp. 33-67.

O’Brien, Kevin and Lianjiang Li. “The Struggle over VillageElections,” in Roderick MacFarquhar and Merle Goldman,eds., The Paradox of China’s Reform. Cambridge, Mass.:Harvard University Press, 1999, pp. 129-44.

Oi, Jean and Scott Rozelle. “Elections and Power: the Locus ofDecision-Making in Chinese Villages,” China Quarterly, vol.


Documents

ELECTIONS, GOVERNANCE, AND ACCOUNTABILITY IN · PDF filethe Chinese are culturally unfit for democracy. ... Governance, and Accountability in Rural China 127 ... Elections, Governance,